HSC President's Essay
A Perspective on Health System Change in 1999
By Paul B. Ginsburg, President, HSC
ach year, I take this opportunity to reflect back on key developments in
health care financing and delivery. When I look back at 1999, what strikes me
most is not the particular actions taken, but the extent to which important
ideas were introduced and discussions initiated in both public policy arenas
and the private sector. It was a year of laying the groundwork for potentially
important future developments. The events of the past year remind me that despite
the dramatic speedup in information dissemination and policy proposal activity,
the time needed to develop ideas, and reach the consensus to implement them
at the national level, is still measured in years rather than months.
- PUBLIC POLICY TO EXPAND INSURANCE COVERAGE
In 1999, this issue received unprecedented attention in the presidential primary campaigns, and new ideas were introduced into the debate from across the political spectrum. - RESPONSE TO RAPIDLY INCREASING EXPENDITURES FOR PRESCRIPTION DRUGS
Employers and health plans are developing mechanisms aimed at consumers so they will economize in this area. Increasing evidence of Medicare beneficiaries' difficulty in paying for prescription drugs launched a substantial discussion about how the federal government should design and administer such a benefit. - RETHINKING MANAGED CARE
Last year saw the continuing backlash against many elements of managed care and the beginning of what may be a significant withdrawal of direct controls on physician behavior. This rethinking of managed care is in response to consumer preferences and to the threat and reality of patient protection legislation at both state and federal levels. - PATIENT SAFETY
The Institute of Medicine's report on patient safety has provided impetus to public policy makers, large employers and those who lead health care delivery systems to find ways to reduce errors.1
EXPANDING INSURANCE COVERAGE
any involved in health policy and private organizations
are worried about increasing rates of uninsured persons in the face of an unprecedented
strong economy. From 1993 to 1998, the number of nonelderly, uninsured persons
increased by 13 percent.2 At the same time,
the proportion of employed adults increased by 9 percent, and per capita income
increased by 13 percent, after adjustment for inflation.
Historically, an improving economy has led to more people obtaining coverage. HSC research examining the rate at which employees enroll in plans offered by employers highlighted a worrisome problem. Twenty percent of those uninsured in 1997 failed to enroll in offered plans, mostly due to cost.3 Large premium increases predicted for 2000 and 2001, and likely continuation of a trend of employers paying a smaller portion of the premium for family coverage, may exacerbate the coverage problem.
Interest in policies to address the problem of the uninsured was strong in 1999. Many influential trade and professional associations developed proposals and attempted informally to find common ground. A January 2000 conference showed a remarkable degree of congruence among organizations representing distinct stakeholders.4 This level of consensus reflects an emphasis on political feasibility and stakeholders having similar assessments of what could be achieved in the near term.
Interest in the issue in the current presidential campaign is substantial. Proposals to expand insurance coverage turned out to be the leading issue in the Democratic primaries. Vice President Gore developed a plan involving expansions of Medicaid and the State Children's Health Insurance Program (SCHIP) that emphasized targeting public funds toward low-income, uninsured persons. Parents with low incomes would also be eligible to obtain coverage for themselves through SCHIP. HSC research has revealed an important decline in coverage for low-income parents between 1997 and 1999.5 Governor Bush's discussion of health care has mentioned a tax credit for low-income workers without access to employment-based coverage and expansion of SCHIP.
Senator Bradley also introduced a number of new ideas into the debate. His tax credit proposal for those with low incomes involved obtaining health insurance through the Federal Employees Health Benefits Program. Those eligible for Medicaid would also obtain private coverage through this mechanism. Parents, with government assistance for those who are low income, would be mandated to obtain coverage for their children.
There appears to be agreement that the federal government should take incremental steps to help low-income persons obtain insurance. There is conflict, however, over whether the principal tool should be expansions of federal and state financing programs or tax subsidies to individuals to purchase private insurance. But with government programs using private plans to provide coverage (e.g., Medicaid managed care), and tax credit proposals needing reforms of the individual insurance market to succeed, divergent ideas may converge when working through the nuts and bolts of mechanisms to blend the roles of the public and private sectors.
RESPONSE TO GROWTH IN PRESCRIPTION DRUG SPENDING
ntil recently, prescription drug benefits did not get a
lot of attention from most health care decision makers. This began to change
in the mid-1990s, when pharmaceutical spending began increasing at double-digit
rates. Contributing to the spending growth are the accelerated rate at which
new drugs are introduced, consumers enjoying more extensive drug coverage with
their shift to managed care and direct-to-consumer advertising.
With this rapid growth coinciding with historically low rates of increase in spending on hospital and physician services, prescription drugs have reached a high enough percentage of private health insurance spending to get the attention of employers and health plans. Pharmaceutical costs accounted for 13 percent of the cost of a typical health insurance policy in 1998, compared to 8 percent in 1994.6
Employers and health plans have sought to manage pharmaceutical spending through contracting with pharmacy benefit management companies (PBMs). These companies negotiate price discounts from manufacturers and pharmacies and implement both closed formularies and tiered copayments to induce consumers and physicians to favor those products considered most effective and to enhance their ability to gain discounts. For specific chronic conditions, such as diabetes, PBMs are working to educate physicians about effective prescribing.
It is striking how rapidly these management initiatives are taking hold. According to a recent Watson Wyatt Worldwide survey of large employers, 47 percent are planning to move to a three-tier copayment system for drugs, and 46 percent are planning to increase the use of formularies.7 This is likely to increase employers' or plans' pricing leverage with pharmaceutical manufacturers. But employees may not react favorably to changes that increase their cost sharing, especially if there are many situations in which a drug has clinical advantages for certain patients but carries the highest copayment. If experience is a guide, employer use of these cost-cutting approaches could fall short of expectations.
In contrast to rapidly implementing efforts to increase patient cost sharing, less has been done to influence physicians to prescribe more cost-effective drugs. For example, efforts to control utilization- or increase utilization to manage chronic disease more effectively- have focused on one disease at a time, and financial incentives for providers are rarely used. Indeed, physician organizations that seek to be at risk for hospital services as well as professional services have been resisting pharmaceutical spending risk, presumably because important trends in spending are outside their control. Overall, the development of mechanisms to control the growth of pharmaceutical spending resembles the evolution of managed care techniques to control spending on hospital and physician services, but it began much later.
Rising prescription drug spending has increased policy makers' concern about affordability for Medicare beneficiaries, which- in contrast to private insurers- did not add these benefits years ago. This has led to proposals for a Medicare prescription drug benefit by President Clinton and both Republican and bipartisan groups in the Congress. (An irony of the difficulty of predicting the policy agenda is that four years ago it appeared that widespread Medicare prescription drug coverage would result from the rapid growth of Medicare+Choice plans.)
At this point, the differences among Medicare drug benefit proposals are substantial. Some are fundamental policy differences that have been debated for decades- for example, should the benefit be limited to those with low incomes or should it be for all seniors? But other issues are new to the policy community, such as how to use pharmacy benefit managers in the program.8
Perhaps the most contentious issue concerns the prices paid for pharmaceuticals. With the pharmaceutical industry making clear its adamant opposition to price controls, all of the major proposals have disavowed such a mechanism. Some have proposed a "most favored nation" mechanism similar to that used in Medicaid. Many have proposed the use of PBMs in the traditional Medicare program to negotiate prices with pharmacies and manufacturers and to provide information services to improve drug use effectiveness. If the Medicare program were to replicate the experience of employers and private health plans, the program would contract with a single PBM for each region.
But the private sector experience might be difficult to replicate in Medicare. With Medicare beneficiaries unable to obtain coverage elsewhere, would policy makers tolerate a situation in which a significant minority of individual beneficiaries was unhappy with the practices of a PBM that was meeting its contractual objectives? Would it be acceptable to include a well-known drug in the formulary in one region, but not in another? Would a single PBM representing so much prescription volume be a de facto implementer of price controls?
These issues are leading some to develop models of multiple PBMs serving Medicare in each region. But these models will be more difficult to design and implement because of the absence of private sector experience to draw on. Employers and health plans do not give enrollees a choice of PBM.9 Also, such a model would ask those beneficiaries who have elected not to enroll in a Medicare+Choice plan to choose a managed care entity for their pharmaceutical coverage. Resolving the mechanism by which prices are set will take time, but the budgetary implications are large enough to require this to be accomplished before a benefit can be enacted.
THE CHANGING FACE OF MANAGED CARE
n previous essays, I have described the backlash against managed
care and the public policy and market responses to that backlash. Some of the
results have included broader choice of providers in managed care plans, more
opportunities for direct access to specialists and better mechanisms for patients
to appeal decisions that deny care.
I do not see any lessening of the backlash. If anything, the debate over patients' bill of rights legislation and state attorney general and class action lawsuits against health plans have intensified the public's negative perceptions of managed care. The return of high rates of premium increase-the earliest employer surveys are showing increases for 2000 in the 9 to 10 percent range-was expected to have made the public rethink the importance of managed care's role in containing costs. But extremely tight labor markets have made employers reluctant to return to more stringent forms of managed care. And prosperity and budget surpluses may be deferring public concerns about rising health care costs.
What we see instead are instances in which managed care plans have pulled back from the mechanisms they had been pursuing to contain costs. Two recent events may turn out to be seminal, signaling an important change in the way health plans manage care.
In November 1999, United Healthcare announced that it would no longer require physicians to obtain authorization to hospitalize a patient, perform or order a major procedure or refer a patient to a specialist. Although physicians are still required to report these actions, United Healthcare would no longer second guess doctors' decisions about appropriate care.10
United Healthcare provided as a rationale for the move the results of its analysis showing that authorization had not been saving money. It claimed that such a high percentage of authorizations were ultimately approved, although often not until appeals by physicians, that the cost of administering the process was exceeding the savings from those services that were ultimately denied. Not explicitly raised were deterrent effects of authorization requirements or the cost of ill will of physicians and their patients whose treatment had been denied. Some critics assert that the announcement exaggerated the magnitude of the change-either to achieve a marketing advantage over competitors or to influence policy makers working on patients' bill of rights legislation.
United Healthcare indicated that some of the administrative resources no longer required for authorization would go to expanded care coordination activities. Among the activities are readmission prevention, disease management and pharmacy management. The company also intends to expand its already significant physician profiling activities to provide additional feedback to physicians on how their practice patterns compare to national norms, and to provide reminders for required patient services. Some anticipate that physician profiling will be used to drop from the network physicians who perform or order a lot of services. Even dropping only a small minority of physicians could influence many other practitioners. Indeed, such a policy could turn out to be even more unpopular among physicians than authorization requirements.
Other managed care companies reacted to United Healthcare's initiative by indicating their intent to move in a similar direction. To the degree that this policy is attractive to physicians and consumers, it could give United an advantage over its competitors. This competitive element could accelerate the degree to which others follow.
A second development was the recent settlement with Aetna Health Plans of a lawsuit brought by the attorney general of Texas against a number of health plans operating in the state. The suit sought to bar the plans from using a range of managed care practices. As part of the settlement, Aetna promised to expand its definition of medical necessity and give treating physicians more authority to determine necessary care. It also promised to stop requiring primary care physicians with small numbers of Aetna patients to be paid on the basis of capitation.
The latter provision may reflect health plans' declining interest in capitation. Use of capitation to pay for physician and other services has not been growing to the degree that had been widely expected three or four years ago.11 Physician enthusiasm for capitation has fallen in response to disappointing financial returns. Health plan enthusiasm has also waned because of the problems many provider organizations have had in managing in this environment.
These developments, combined with others such as more direct access to specialists and broader networks, suggest that managed care may be evolving toward a model that places less emphasis on traditional methods of cost control. Although health plans claim that cost containment is not being sacrificed, the fact that they are reacting to strong pressure to reduce controls from both their customers and policy makers means this is unlikely to be the case.
What is uncertain is whether this period of less management will be brief or lengthy, and the shape of the next generation of strategies to control costs. The magnitude of cost increases experienced over the next two years will play a major role in shaping the outcome. Should costs rise more rapidly, employers will have to decide whether to attempt to shift them to employees or return to tighter management. Policy makers' perspectives are likely to be influenced by increases in outlays for Medicare and Medicaid and in the number of uninsured due to the affordability of coverage.
Information technology is likely to play a more important role in the next generation of managed care, but the nature of that role is still unclear. One possible direction is the care coordination that United Healthcare is promoting, which makes use of information technology to support physicians' practicing more effectively. But past disappointments in progress toward this goal lead me to be skeptical about its near-term promise. Information technology may facilitate the use of consumer choice as a cost-containment strategy. For example, the Internet could facilitate offering those enrolled in a health plan different networks of providers at different prices. Assessing the changing face of managed care is a priority in HSC's third round of site visits, which began in June 2000.
MEDICAL ERRORS AND PATIENT SAFETY
n November 1999, The Institute of Medicine (IOM) released
its study, To Err Is Human: Building a Safer Health System,12
to enormous media attention. The study highlighted the degree to which medical
errors injured patients and explained that most were not the result of carelessness
by individual practitioners but resulted from shortcomings in systems. It also
made a series of specific recommendations designed to stimulate the creation
of safety systems within health care organizations. These included identifying
and learning from errors on the basis of mandatory reporting (with public disclosure)
of adverse events that result in death or serious harm and voluntary reporting
(with confidentiality protected) of other errors. The study drew heavily on
the experience of other industries- for example, airlines, in which system approaches
have been highly successful in promoting safety.
Many were struck by the degree to which this report gained the attention of the general public, in contrast to earlier major reports on quality of care, such as the President's Advisory Commission on Consumer Protection and Quality in the Health Care Industry.13 This aspect of quality may be easier for the public to grasp. Whatever the reason, federal and state policy makers immediately seized the opportunity provided by the widespread public response and introduced legislation to reduce medical errors.
Like other issues that suddenly become visible to the public, experts have long recognized the significance of medical errors as a source of injury. Over the past several years, leading employers have been concerned about data on errors and have launched initiatives to prod health service providers, usually hospitals, to develop systems to prevent errors-for example, automated order entry systems.14
The question to contemplate is whether the IOM report and the reaction to it will lead to more rapid progress in this area. The report may spur large employers to do more and make more credible their attempts to get providers to develop systems to reduce errors. While some employers comprise an important enough share of the privately insured in a community to be in a position to influence hospitals, most do not. Employers are likely to have more leverage on this issue working through health plans. But recent consumer demands for broad choice of providers have diminished the clout of employers or health plans to push hospitals to invest in systems to reduce medical errors.
Policy makers have a great deal of potential to encourage or discourage efforts to increase patient safety. Funding research on ways to reduce errors would likely be a positive step, and it appears to have consensus support.
Reporting requirements are more complex. Public disclosure of errors would increase provider incentives to invest in systems to reduce errors, but it might also encourage more covering up. The IOM sought a balance between disclosure of the most severe errors (emphasizing accountability) and protecting the confidentiality of reports of less severe errors (emphasizing the opportunities to learn from them). For the latter-the approach used in the airline industry-to work, providers would need strong assurances that the information could not be used as evidence in malpractice suits. A key task for the policy process is to understand this tradeoff: individuals not having access to information on fault, which could be used to sue for compensation for minor injuries, in order to enhance patient safety. Such information could be used for the common good, in that it could be analyzed to figure out how to reduce overall errors.
eflecting on these four developments in 1999, it is striking
how slowly some changes proceed. Those involved in public expansions of coverage
are attempting only incremental change and fighting against a tide of rising
premiums and a trend toward less employer support of family coverage. Responses
to rising prescription drug spending are limited by a lack of experience with
the use of relevant cost control mechanisms. Despite pressure to decrease the
use of many traditional managed care tools, the next generation of care management
tools is still being formulated and is years behind what had been predicted.
And in efforts to draw on successes in other industries in improving safety,
difficult tradeoffs that seem to be unique to health care must be carefully
addressed.
My sense is that recent advances in communication will help both public and private decision makers to start implementing those changes that have garnered consensus more rapidly than they would have in the past. But health care is complex, and given all the powerful entrenched interests, progress is likely to be slow.
Paul B. Ginsburg
PERSPECTIVES
CHARLES N. KAHN III,
HEALTH INSURANCE ASSOCIATION OF AMERICA
"The presidential candidates' proposals to expand health coverage reflect differences in philosophy, but both sides realize it will take new public policy to shrink the rolls of the uninsured. Interestingly, a real common ground is developing among industry and consumer groups about next steps that may help push the policy makers to solutions in 2001.
JOSEPH NEWHOUSE,
HARVARD UNIVERSITY
"Medicare pharmaceutical coverage was barely on the radar screen three years ago, but has quickly shot to the top of the policy agenda. Aside from the usual challenges of enacting an expansion to a public program, this one includes myriad complex implementation issues that must be confronted."
MARGARET O'KANE,
NATIONAL COMMITTEE FOR QUALITY ASSURANCE
"The public does not perceive the relationship between its message to the managed care industry to pull back on management and the problems of health care affordability that will result."
BRUCE BRADLEY,
GENERAL MOTORS CORPORATION
"Reducing medical errors is a big-tent issue if the emphasis can be put on examining the root causes of mistakes and making appropriate changes, as opposed to placing blame."
SOURCES
1 Linda T. Kohn, Janet Corrigan and Molla Donaldson. "To Err Is Human: Building a Safer Health System." Institute of Medicine, National Academy Press (1999).
2 Paul Fronstin. "Sources of Health Insurance and Characteristics of the Uninsured: Analysis of the March 1999 Current Population Survey." Issue Brief, No. 217. Employee Benefit Research Institute (January 2000).
3 Peter J. Cunningham, Elizabeth Schaefer and Christopher Hogan. "Who Declines Employer-Sponsored Health Insurance and Is Uninsured?" Issue Brief, No. 23. Center for Studying Health System Change (November 1999).
4 "RWJF Sets Up Insurance Coverage Initiatives." Advances, Issue 1. The Robert Wood Johnson Foundation (2000).
5 Peter J. Cunningham and Michael H. Park. "Recent Trends in Children's Health Insurance Coverage: No Gains for Low-Income Children." Issue Brief, No. 29. Center for Studying Health System Change (April 2000).
6 Paul B. Ginsburg. "Tracking Health Care Costs: Long-Predicted Upturn Appears." Issue Brief, No. 23. Center for Studying Health System Change (November 1999).
7 "Putting Employees in Charge: A Survey of Employers, Health Care Providers and Health Plans." Fifth Annual Survey Report on Purchasing Value in Health Care 2000. Watson Wyatt Worldwide (2000).
8 The analysis and debate that began in 1999 (see Health Affairs, March/April 2000, for an excellent group of papers) will increase to a substantial degree our understanding of how to effectively provide a benefit.
9 For an excellent discussion of these design issues, see Haiden A. Huskamp et al., "The Medicare Prescription Drug Benefit: How Will the Game Be Played?" Health Affairs (March/April 2000).
10 This change would not affect determinations of whether a service is covered under the plan.
11 Cara Lesser and Paul B. Ginsburg. "Update on the Nation's Health Care System: Results from Tracking Change in 12 Communities." Presented at the HSC conference, Update on the Nation's Health Care System: Results from Tracking in 12 Communities. Washington, DC (November 16, 1999).
13 "Quality First: Better Health Care for All Americans." Washington, DC: U.S. Government Printing Office (1999).
14 "White Paper on Value-based Purchasing" (unpublished manuscript). Leapfrog Group (January 25, 2000).
IN PREPARING THIS ESSAY, I BENEFITED FROM HELPFUL COMMENTS FROM BRUCE BRADLEY, JANET CORRIGAN, JACK EBELER, ANN GREINER, JOE NEWHOUSE, CHIP KAHN, PETER KEMPER, DAVID MUNDEL AND PEGGY O'KANE.
